I just want to chime in and say this is exactly how you should use this Ai to code.

dont browse the web Facebook or any modern Google/iPhone Store apps.

The internet used to be a space for weird geeky hobbyists that more traditional plebs couldn't access or couldn't be bothered to fuck with. Now it's still that, but it has a bunch of shit for the rubes, too.

At some point, I feel like I'm talking to someone who says "I fucking hate Florida. Every time I go, I spend a week at Disney World and it's expensive and awful and loud and stupid." And here I am, out in the Keys, working on my tan and fishing and hiking and hooking up with cuties, having no problems whatsoever.

We are not at the end of the road. We are not at the beginning of the end of the road. We are not at the end of the beginning.

I definitely get the impulse to doom. And I'm as prone to it as anyone. But when I look at crypto and AI, all I can see is the same analog fuck-ups made in prior generations. Beanie Babies and Labubus didn't ruin the stuffed animal industry. The Delorean and the Hummer didn't ruin the automotive industry. The Great Depression of 1932 didn't ruin the financial sector.

Plenty of things to be excited about in software and tech that lives entirely outside the cloistered hype-beast market. Raspberry Pis, 3D printers, 3nm chipsets built with ultraviolet lithography, solid state drives, lithium and sodium ion batteries with incredibly recharge rates, gorilla glass and carbon fiber, 5G+ radios, full voice recognition, self-piloting vehicles.

How is none of this thrilling? Hell, even just the advent of coding pipelines that can take a project from a funky coding idea to a deliverable feature in a few keystrokes is such a huge step forward in development. I can't hate the sales goons pushing junk when I'm so immersed in all the novel innovative applications of technology I've been watching bud itself up from the ground for the last 40 years.

Even LLMs on their face are such a novel application of graph theory. You can do so much cool stuff off a second hand laptop today. It's an exciting new frontier.

Everything is a scam. It feels like no engineers exist any longer. It’s not even just tech it’s everything. I bought a jar from a major retailer and the lids don’t fit right on them.

It’s pretty obvious capitalism has run its course. What i don’t get is the amount of people helping destroy the world they have to live in or worse—they are having kids in a world they are actively making shitty for six figures.

Then they use their kids as the excuse to make the world worse for everyone else! “I can’t afford not to!”

Well, what you actually can’t afford is to keep doing this. Quit working for these public corps. Sacrifice everything to not work for them. I promise you’ll be better off after some pain. Coming from someone with no actual education that made it out you college grads can do it. I believe in you.

And your kids don’t need large Christmas’s and a new car at 16 with private school. They do need a functioning world.

I can honestly say that I dread the future of tech, and the goat farm thing sounds really appealing.

I went to a local store the other day and saw 2 new flock cameras that went up. There was one person manning 8 self checkout stations, and there was a camera watching me scan and bag my own stuff while displaying a live feed of me. I hopped in my car, which automatically turned on GPS and then my phone started giving me "helpful" suggestions like " have you tried out this restaurant nearby?" and "it's been a while since you've been to the pet store to get stuff". I suddenly felt like a boiled frog, because I don't remember turning any of that shit on, or being notified about it being turned on, or seeing info about new surveillance cameras going up in the community. At this point, I want to buy some land in the middle of fucking nowhere and disconnect from all that shit, and I legitimately had a low-level panic about all of it.

Sure, AI can automate the work, but instead of maintaining quality, said quality of work is also taking a nosedive in tandem with the quality of materials.

Knuth wrote about AI solving one of his 30 year old problems:

https://www-cs-faculty.stanford.edu/~knuth/papers/claude-cycles.pdf

So it can be good in the right hands.

OHHHH. In that case BARTENDER this person's next drink is on me.

I was alluding to the wider web being fucking ass now. It has nothing to do with Linux. Thats why staying on your machine is a nostalgia trip to the good times.

What? Sounds like the last time you tried it was a decade ago.

That sucks, that your passion for your job got eroded to this point. The z scale train in a suitcase sounds like a really cool project. Share some updates along the way.

I'm still cool with tech, but only if there is a radical shift in how it's being used and integrated into society as a whole. In my view, people are using tech as a crutch or substitute to supplement the limitations of biology. IMHO that is the wrong way to go about it.

The disgusting weaknesses of the flesh should not be supported by technology, but should instead be supplanted by it. The purity of the machine is not a stepping stone to some higher enlightened state, it IS the enlightened state. At the same time, we must shun the heresy of the Abominable Intelligence and instead wed the purity of true thought and knowledge to a vessel that is capable of fully encompassing its divinity.

The crude biomass which is generally held to be a temple is nothing more than the result of random trial and error. Inevitably it will fail. In that moment, only those who aspire to the perfection of the machine and the blessed strength of steel will receive salvation.

I serve the Omnissiah.

It's not only IT. I'm pretty sure we've peaked. We don't try to produce good products or services anymore, just ones that can get quick returns. Get customers and then make it worse, for the shareholders!

I still have products from >10 years ago which are great and work, but you can buy less and less things like that. Even in sectors that were super reliable.

I'd push back on your point here with a few things:

The primary one being: the code doesn't need to be perfect or even above average -- average is perfectly fine. The idea here is comparing the AI to a human, not to perfection. I see this constantly with AI and I find it a bit disingenuous.

I do truly believe what I said above will be possible within my career (I'm in my mid 30s), but it's not really what I'm worried about right now. I think the current code I see being generated is generally "good enough". I'm not comparing it to perfect: I'm comparing it to people.

I read a comment once that still rings true - “Hallucinations” are a misnomer. Everything an LLM puts out is a hallucination; it’s just that a lot of the time, it happens to be accurate. Eliminating that last percentage of inaccurate hallucinations is going to be nearly impossible.

I don't see any reason you have to remove all hallucinations to get a good tool for autonomous development: humans aren't perfect either. We compensate for that with processes and checking each others work, but plenty still falls through the cracks.

LLMs also have no understanding of context outside the immediate. Satire is completely opaque to them. Sarcasm is lost on them, by and large. And they have no way to differentiate between good and bad output. Or good and bad input, for that matter. Joke pseudocode is just as valid in their training corpus as dire warnings about insecure code.

Have you seen output in which satirical code is actually included? I'm well aware of things like https://www.anthropic.com/research/small-samples-poison and the potential here. And do you not believe that either (a) these types of trivial issues would be caught by a person whose job was just to audit output or even (b) this type of issue could be caught by specially trained domain limited AIs designed to check output?

To your point then: what are your thoughts on this project? https://github.com/anthropics/claudes-c-compiler I'm not particularly interested in this use case right now but it seems more in line with what you're interested in.

I think it shows a lot of limitations but also a lot of potential. I don't personally think the AI needs to get the code perfect on the first go -- it has to be compared to humans and we definitely don't do that.

I really really dislike the way it’s being sold as a solution for things it’s in no way a solution for.

Yes, of course. I think it's important to look passed the blowhards and think about what it's actually doing: that is the perspective I'm trying to talk about this from.

Wahoo! Best of luck!

Wow, that group +s seems exactly what I'm looking for! That actually looks like the clean solution I was looking for. I'll test it out and report back, I'll have to wait on Monday for the colleagues to be back in the server, but it seems very promising.

Thank you very much!

None of these programs allow overriding Linux permissions. You can not recline/rsync in another user directory. You can not sftp/scp in another user directory. My problem is not about transferring data across different systems, but rather accessing data on one system through different users. All users should be able to read and modify the files.

I didn't intend and don't think the stick bit stuff will or could be a complete solution for you. You've got some oddly specific and kinda cruddy restrictions that you've got to workaround and when they get that nonsensical one ends up solidly in "cruddy hack" territory.

From the article:

group + s (pecial)

Commonly noted as SGID, this special permission has a couple of functions:

If set on a file, it allows the file to be executed as the group that owns the file (similar to SUID) If set on a directory, any files created in the directory will have their group ownership set to that of the directory owner

You could run something like https://pypi.org/project/uploadserver/ in screen or run a cron every minute that just recursively sets the correct permissions.

I thought sticky bits were used to allow other users to edit files but not delete them. Do they also allow inheriting the parent directory permissions?

It's just default Linux permissions. People who get access to the HPC through the same institution are placed in the same group. As you may know scientific collaboration is quite important. Indeed when collaborating we sign all the necessary paperwork, but that does not translate to the HPC administrators who are part of a separate institution. To request a separate group you have to contact the HPC institution, they will have to contact the institutions involved. Those institutions will have to check that NDAs are already in place. If the NDAs are in place they will have to check that the data to be shared is actually covered by the project. I will have to fill a bunch of paperwork. This will be sent to an external auditor to check that everything is correct, and then everything goes back in that chain.

I already waste way too much of my time on paperwork. Worst case scenario is a collaborator leaks some data which will be published publicly in a few months anyway. And those collaborators will have access to such data anyway, just through other less comfortable means.

Maybe some sticky bit https://www.redhat.com/en/blog/suid-sgid-sticky-bit

There’s a reason for this governance and you’re putting your whole team at risk trying to do this yourself

Here's someone that solved this by monitoring the directory using inotifywait, but based on the restrictions you already mentioned I'm assuming you can't install packages or set up root daemons, correct?

https://bbs.archlinux.org/viewtopic.php?id=280937

Edit: CallMeAI beat me with this exact same answer by 15 minutes.

We can not setup a common group, no way we get root privileges. A cron job would not work either: it is a cluster with many nodes, of which many login nodes. Cron jobs do not work on such systems.

A share_cp script would in fact be a good solution, I may try that and see if people pick it up.